Multi-illumination-interfered Neural Holography with Expanded Eyebox
Xinxing Xia^1^, Pengfei Mi^1^, Yiqing Tao^1^, Xiangyu Meng^2^, Wenbin Zhou^2^, Yingqie Yu^1^, Pengfei Shen, Yifan (Evan) Peng^2^*
1Shanghai University
2The University of Hong Kong
IEEE Transactions on Visualization and Computer Graphics (ISMAR), 2025
Jan 15, 2025
Abstract
Multi-illumination-interfered neural holography represents a significant advancement in holographic display technology. We present a comprehensive approach that leverages multiple illumination sources with carefully designed interference patterns to dramatically expand the eyebox of holographic displays. Our neural optimization framework learns to generate holograms that maintain high visual quality across a much wider viewing angle than traditional methods. Through extensive experiments and theoretical analysis, we demonstrate substantial improvements in both eyebox size and image quality, making practical holographic displays more feasible for real-world applications.
Summary
We introduce a novel approach for neural holography that uses multi-illumination interference to achieve expanded eyebox in holographic displays. Our method addresses the fundamental challenge of limited viewing angle in traditional holographic systems through advanced neural optimization techniques.
Overview
This work presents a breakthrough in neural holography by introducing multi-illumination interference techniques that significantly expand the eyebox of holographic displays. The expanded eyebox addresses one of the most critical limitations in current holographic display systems.
Key Contributions
- Novel multi-illumination interference framework for neural holography
- Significant expansion of holographic display eyebox
- Advanced neural optimization techniques for hologram generation
- Comprehensive evaluation demonstrating improved viewing angles and image quality
Technical Approach
Our method combines multiple illumination sources with neural optimization to create holograms that maintain visual quality across expanded viewing angles. The interference patterns are carefully designed to maximize the effective eyebox while preserving image fidelity.
Citation
@article{xia2025multiillumination,
title={Multi-illumination-interfered Neural Holography with Expanded Eyebox},
author={Xinxing Xia and Pengfei Mi and Yiqing Tao and Xiangyu Meng and Wenbin Zhou and Yingqie Yu and Pengfei Shen and Yifan Peng},
journal={IEEE Transactions on Visualization and Computer Graphics},
year={2025},
publisher={IEEE}
}